Non-stress test fetal heart rate strip interpretation — baseline, variability, accelerations & reactivity
The non-stress test (NST) is a widely used antepartum test of fetal well-being that records fetal heart rate (FHR) and uterine activity externally, without inducing contractions. It exploits a simple physiologic principle: an adequately oxygenated fetus with an intact autonomic nervous system and central nervous system will intermittently accelerate its heart rate in response to movement. The first step in reading any strip is establishing the baseline rate.
The NST is typically performed starting around 32–34 weeks (earlier in high-risk pregnancies) in patients with conditions that raise concern for uteroplacental insufficiency: hypertensive disorders of pregnancy, diabetes, fetal growth restriction, decreased fetal movement, post-term pregnancy, oligohydramnios, and multiple gestation, among others.
The test is grounded in fetal autonomic physiology: a well-oxygenated fetal brainstem and autonomic nervous system produce spontaneous heart-rate accelerations, usually coupled to fetal movement, mediated by sympathetic outflow. Hypoxemia, acidemia, CNS depression, or immaturity blunt this reflex — so the pattern of accelerations (and variability) is used as an indirect, non-invasive readout of fetal oxygenation and neurologic integrity at the moment of testing.
Baseline FHR is the approximate mean rate rounded to the nearest 5 bpm during a 10-minute window, excluding periods of marked variability, accelerations, and decelerations. At least 2 minutes of identifiable baseline segments (not necessarily contiguous) are required in any 10-minute window to assign a baseline; if the baseline is truly indeterminate, the prior 10-minute window is used for reference.
Baseline is reassessed as the tracing continues — it is not a single fixed number for the whole test, but a rolling determination.
Tachycardia (>160 bpm): maternal fever or chorioamnionitis, early/compensated fetal hypoxemia, fetal tachyarrhythmia, maternal hyperthyroidism, beta-sympathomimetic or anticholinergic medications, fetal anemia, prematurity.
Bradycardia (<110 bpm): fetal congenital heart block (e.g. associated with maternal anti-Ro/SSA antibodies), prolonged cord compression or occult prolapse, maternal hypotension or hypothermia, local anesthetic toxicity, and, when severe or evolving, fetal hypoxemia — bradycardia in that setting is an ominous rather than reassuring finding.
Baseline variability describes the beat-to-beat and short-term fluctuation of the FHR around the baseline. It is generated by the continuous, competing input of the sympathetic and parasympathetic branches of the fetal autonomic nervous system on the sinoatrial node, and most clinicians consider it the single most reassuring feature on a tracing when present in the moderate range.
Variability is generated by rapid, opposing autonomic input: sympathetic tone speeds the heart, parasympathetic (vagal) tone slows it, and their beat-to-beat interplay produces the irregular, jagged appearance of a healthy trace. Because this circuitry requires an intact brainstem, cortex, and cardiac conduction system, variability is used as a real-time marker of fetal central nervous system integrity and, indirectly, of adequate cerebral oxygenation.
Unlike accelerations, which are episodic, variability is present essentially continuously on a healthy tracing — that continuity is part of what makes it such a robust reassurance signal.
Moderate variability (6–25 bpm) is considered strongly predictive of a non-acidemic fetus at the time of testing and is the variability category built into the definition of a reassuring tracing.
Moderate variability is reassuring on its own. Absent or minimal variability is concerning, particularly when accompanied by decelerations or a non-reactive pattern, and raises concern for fetal acidemia, CNS depression, or a compromised autonomic reflex arc — though it is frequently caused by benign, transient states as well (see below).
Marked variability (>25 bpm, sometimes described as a "saltatory" pattern) has less well-defined isolated significance; it can be seen with acute partial hypoxic stress, fetal stimulation, or cord compression, and when persistent or accompanied by decelerations it warrants closer evaluation.
The most frequent cause of reduced variability on an otherwise healthy fetus is the fetal sleep cycle — quiet sleep can transiently minimize variability and reactivity for 20–40 minutes, which is why extending the monitoring period is the first response to a non-reactive strip.
Other causes: prematurity (before ~32 weeks the autonomic system is less mature and both variability and reactivity are physiologically reduced), maternal medications that cross the placenta (opioids, benzodiazepines, magnesium sulfate, general anesthetics, betamethasone in the 24–72 hours after administration), fetal acidemia or hypoxemia, and rarely fetal anomalies affecting the CNS (e.g. anencephaly).
An acceleration is an abrupt (onset to peak in under 30 seconds), visually apparent increase in FHR above the most recently determined baseline. Accelerations are usually associated with fetal movement and are the central building block of the reactive/non-reactive classification that gives the non-stress test its name.
At ≥32 weeks gestation, an acceleration must rise ≥15 bpm above baseline and remain at or above that level for ≥15 seconds, but less than 2 minutes, measured from the most recently determined baseline (not from the trough of a prior deceleration). Before 32 weeks, the immature autonomic reflex produces smaller, shorter accelerations, so the threshold is relaxed to ≥10 bpm for ≥10 seconds.
An acceleration lasting ≥2 minutes but <10 minutes is termed a prolonged acceleration; ≥10 minutes redefines the baseline itself.
A reactive (reassuring) NST is defined as ≥2 qualifying accelerations occurring within any 20-minute window. Because fetal quiet sleep cycles typically last 20–40 minutes, a tracing that is not yet reactive at 20 minutes is not necessarily abnormal — professional guidelines support extending observation to 40 minutes before labeling the test non-reactive, to allow the fetus to cycle into an active state.
Reactive = ≥2 accelerations of ≥15 bpm for ≥15 seconds within 20 minutes (≥32 weeks); 10 bpm / 10 seconds before 32 weeks. Extend to 40 minutes before calling a strip non-reactive.
A qualifying acceleration demonstrates an intact chain of physiology: fetal movement, an aroused autonomic state, and an appropriately responsive sympathetically-mediated heart rate rise — all of which require adequate oxygenation and an unimpaired central and autonomic nervous system at that moment. A reactive tracing is strongly, though not perfectly, associated with the absence of significant fetal acidemia at the time of the test and carries a very low associated risk of stillbirth in the following week.
The NST is designed to observe spontaneous FHR behavior without induced contractions, and reactivity is defined purely by baseline and accelerations. Decelerations are not part of the formal reactive/non-reactive definition — but when they occur during an NST they are never simply disregarded, and recognizing their pattern is essential to safe interpretation.
Because the NST records spontaneous, unstimulated uterine activity, contractions are usually infrequent or absent, so contraction-linked decelerations (early, late) are uncommon by design — the test simply is not built to provoke them. Spontaneous variable decelerations can still occur from transient cord compression (e.g. oligohydramnios, cord position) independent of contractions. Any deceleration pattern noted during an NST changes its interpretation and management, even though it does not, by itself, define reactivity.
Early decelerations: gradual onset/offset, nadir coincides with the contraction peak — mirror image of the contraction curve — caused by fetal head compression and vagal reflex; considered physiologic and benign.
Variable decelerations: abrupt onset and offset (<30 seconds to nadir), variable in shape, depth and timing relative to contractions — caused by umbilical cord compression; significance depends on depth, duration, and recovery.
Late decelerations: gradual onset/offset, nadir occurs after the contraction peak (delayed relative to the contraction) — reflect transient fetal hypoxemia from uteroplacental insufficiency triggering a chemoreceptor-mediated reflex; recurrent late decelerations are non-reassuring and warrant prompt evaluation.
The illustrative dips overlaid on this stage are shaped and timed like late decelerations — gradual, delayed relative to the toco contraction peak — to demonstrate a non-reassuring pattern. They are synthetic teaching overlays, not part of the reactive/non-reactive scoring shown in the metrics panel.
Isolated, mild variable decelerations with a normal baseline, moderate variability, and accelerations present are usually not concerning. Recurrent, deep, or late decelerations — especially paired with minimal/absent variability or a non-reactive pattern — prompt closer evaluation: assessment for cord compression or oligohydramnios, maternal repositioning and oxygen, and consideration of a biophysical profile (BPP), contraction stress test (CST), or expedited delivery depending on gestational age and overall clinical context.
A useful NST interpretation combines all three signals — baseline, variability, and accelerations — into one overall call, then routes non-reassuring findings toward a clear next step. This final stage synthesizes the tracing into a single reactive/non-reactive verdict, the same synthesis a clinician performs at the bedside.
Reactive NST: baseline in the normal range (110–160 bpm), moderate variability present, and ≥2 qualifying accelerations (≥15 bpm for ≥15 sec at ≥32 weeks; ≥10 bpm for ≥10 sec if <32 weeks) within a 20-minute window (extendable to 40 minutes).
Non-reactive NST: the above criteria are not met after an adequate observation period — most commonly insufficient accelerations, but a non-reassuring baseline (tachycardia/bradycardia), or absent/minimal variability contribute to the overall picture as well.
Reactive = normal baseline + moderate variability + ≥2 qualifying accelerations in 20 (up to 40) minutes. Any one criterion failing does not automatically mean fetal compromise — it means "look further," starting with time and stimulation before escalating.
1. Extend monitoring to 40 minutes total to allow the fetus to cycle out of quiet sleep. 2. Vibroacoustic stimulation (VAS): a brief acoustic stimulus applied to the maternal abdomen to arouse the fetus and elicit an acceleration — shortens testing time and reduces the false non-reactive rate substantially. 3. If still non-reactive, proceed to a biophysical profile (BPP, which incorporates the NST with ultrasound assessment of tone, movement, breathing, and amniotic fluid) or a contraction stress test (CST), and integrate the result with the overall clinical picture (gestational age, indication for testing, other risk factors). 4. Persistently non-reassuring findings, especially with a low BPP score or abnormal CST, may prompt delivery depending on gestational age and risk-benefit assessment.
The NST has a high false non-reactive ("false-positive") rate — many non-reactive tracings occur in perfectly healthy fetuses, most often because of the fetal sleep cycle. Prematurity (<32 weeks) further increases the false non-reactive rate because the autonomic reflex arc is not yet fully mature.
Maternal medications and states that can transiently blunt variability/reactivity: opioid analgesics, sedatives and benzodiazepines, magnesium sulfate, antenatal corticosteroids (transiently, 24–72h), and maternal smoking immediately before testing.
False negatives (fetal compromise despite a reactive test) are uncommon but possible, since the NST is a snapshot in time — it reflects fetal status only at the moment of testing and does not exclude an evolving process. For this reason NST is typically repeated on a schedule (e.g. weekly or twice weekly) in ongoing surveillance rather than relied upon as a single isolated test.